Imagine watching an embryo form in real time, each cell dividing, moving, and finding its place. Today’s advanced microscopes can capture these frames—but turning them into accurate, usable maps of cellular trajectories has long been a very difficult computational problem in biology.
With the release of two new tools from the Chan Zuckerberg Biohub San Francisco and collaborators, that problem may finally be getting easier. In Nature Methods, scientists reported: “Ultrack: pushing the limits of cell tracking across biological scales” and “inTRACKtive: a web-based tool for interactive cell tracking visualization.”
Together, the tools may change how researchers follow cells across space and time, from development to disease. “Tracking live cells across two-dimensional, three-dimensional (3D), and multichannel time-lapse recordings is crucial for understanding tissue-scale biological processes,” the Ultrack paper noted. Yet segmentation errors and the need to retrain deep learning models on each new dataset have slowed adoption in many labs.
Ultrack addresses these bottlenecks by simultaneously solving the tasks of segmentation and linking, rather than treating them as separate steps. This integrated approach makes the algorithm faster, more accurate, and scalable from small cultures to terabyte-scale datasets of whole embryos. Its performance, however, still depends on the quality of the foreground and contour maps; widespread segmentation errors or unusually large cell movements between frames can reduce accuracy unless additional registration is applied, reported the authors.
“It’s easy to do tracking in 2D or on a few cells, but Ultrack pushes the limits on very hard scenarios, like 3D or full embryos,” said Loïc Royer, PhD, director of imaging AI at the San Francisco Biohub and senior author of the study. “It’s very fast and scales well, but also has a lot of practical features to make it easy to use.”
The team evaluated Ultrack on zebrafish development, reconstructing entire trajectories of neuromast cells, and compared its performance to benchmarks in the international Cell Tracking Challenge. The algorithm achieved superior or comparable performance, the researchers said.
When asked about Ultrack’s uses, Royer told GEN, “Pharma and biotech companies interested in leveraging modern imaging technologies, especially 3D imaging of tissues or model organisms, can leverage Ultrack to extract cellular trajectories and make sense of their data.”
But generating the data is only part of the challenge. As the authors of the companion paper explained, “While these imaging datasets hold the potential for biological insights, accessing and interacting with such complex data typically demands highly specialized technical expertise and substantial computational resources.”
That’s where inTRACKtive comes in. The web-based tool turns massive cell tracking datasets into an explorable, interactive 3D environment—accessible from any browser, no installation required. Researchers can rotate embryos, select cells or trace lineages, color cells by attributes, and even share exact views through a simple link.
To demonstrate its power, the team launched the Virtual Embryo Zoo, an online collection of whole-embryo datasets from six model organisms, including zebrafish, mouse, and C. elegans. “We encourage researchers to contribute to the Virtual Embryo Zoo by submitting their own whole-embryo datasets from other species,” said first author Teun Huijben, PhD.
By lowering technical barriers, the tools could also broaden adoption outside of developmental biology. Potential applications range from cancer research to regenerative medicine, “…offering new insights into cellular behavior and tissue dynamics,” reported the Ultrack researchers.
Regarding pairing with other tools or platforms in the future, Royer added, “…we are actively working on bridging modalities, and both Ultrack and inTRACKtive are helping us to do that. While Ultrack performs extremely well, we are still far from where we want to be in terms of accuracy, but we have good hope that in the next few years, we will be making significant progress toward near-perfect cell tracking.”
With Ultrack and inTRACKtive, what was once a painstaking process of stitching together cell paths may soon become an interactive, collaborative, and scalable enterprise—one that could reshape how biology and biotechnology see cells in motion.
The post Cell Tracking Transformed with Scalable AI and Interactive Visualization appeared first on GEN – Genetic Engineering and Biotechnology News.
